[fdo-internals] Question about FDOOGRProvider

Jackie Ng jumpinjackie at gmail.com
Fri Mar 28 20:50:36 PDT 2014

If a spatial context is static, an extent *must* be returned if asked for,
even if it is a completely fabricated one. If such a fabricated extent is
returned, it means the metadata defining such bounds doesn't exist or the
actual logic to identify the bounds isn't implemented.

You may have noticed this behavior in the GDAL provider in the past. Of
course, that was due to a bug (see: http://trac.osgeo.org/fdo/ticket/493),
but the +/- 2000000 extent was always the last-resort extent used if the
true bounds could not be identified.

This may not be actual spec'd behavior for this spatial context API, but
it's how I imagine most FDO client applications are behaving when they want
an extent.

Now for the PostGIS FDO provider, I guess actual extent computation (whether
by ST_Extent or ST_EstimatedExtent) wasn't implemented which is why a +/-
2000000 extent was returned.

- Jackie

View this message in context: http://osgeo-org.1560.x6.nabble.com/Question-about-FDOOGRProvider-tp5131270p5131984.html
Sent from the FDO Internals mailing list archive at Nabble.com.

More information about the fdo-internals mailing list